GSM8KactiveMathematics
GSM8K: Grade School Math 8K
Metric: Accuracy (higher is better)Introduced: 2021
8,500 grade school math word problems requiring 2–8 steps of arithmetic reasoning. Widely used for evaluating chain-of-thought reasoning; top models now near 100%.
Models ranked
10
tracked on this benchmark
Score band
95.7 – 80.6
best → lowest tracked
Snapshot trend
+2.37
May 28 → Jun 25 · 1 models
Leaderboard
Tracked models ranked by Accuracy (higher is better).
#Model variant and provenanceRelative to leaderScore
1
Llama 3.1 405B
95.7Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
2
Granite 4.1 30B
94.2Version: 8-shot, instruct model (accuracy)Harness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
3
Llama 3 70B
93.0Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
4
Granite 4.1 8B
92.5Version: 8-shot, instruct model (accuracy)Harness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
5
Nemotron-Labs TwoTower 30B-A3B Base
90.1Version: 8-shot, accuracy; default TwoTower diffusion decoding at confidence_threshold=0.8, block_size=16, BF16 on 2xH100; evaluator/harness not published on the model cardHarness: Not recordedEvaluator: Not recordedObserved: Jun 25, 2026Confidence: Not recordedSource
6
Gemma 2 27B
89.7Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
7
Granite 4.1 3B
86.9Version: 8-shot, instruct model (accuracy)Harness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
8
Phi-3 Mini 128K
84.9Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
9
Qwen2-7B
82.7Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
10
Llama 3 8B Instruct
80.6Version: Not recordedHarness: Not recordedEvaluator: Not recordedObserved: May 28, 2026Confidence: Not recordedSource
How to read this benchmark
This benchmark scores models where higher is better. Use scores for directional filtering and shortlisting, not universal quality ranking; then validate pricing, context window, provider availability, and fit for your workload.
Trust this score when
- There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
- The model list covers the same version family you can actually deploy today.
- Top candidates overlap with your required routing and feature requirements.
Be cautious when
- There is only one benchmark snapshot or the dataset appears stale.
- The benchmark metric direction is opposite of your decision objective.
- The score difference between options is narrow and likely within implementation variance.
Related benchmarks
Last reviewed: Apr 15, 2026